Video Sending Rate Prediction Based on Communication Logging Database for 5G HetNet

Abstract

Recently research and development on the 5th generation mobile communication system (5G) is actively carried out to launch commercial service in 2020. In 5G, Super high frequency (SHF) band and extremely high frequency (EHF) band are used for realizing high speed and large capacity communication. Furthermore, 5G network will include many wireless LANs (Local Area Networks) as well as cellular network. Therefore, 5G network is composed as the heterogeneous network (HetNet). In 5G HetNet, a user equipment (UE) connects various base stations (BS) according to its movement. Accordingly, it is assumed that the throughput suddenly fluctuates due to the change of radio links. By the way, 5G is expected as a communication means for large capacity video contents such as 4K/8K video and omnidirectional video. However there is a problem that the conventional video transmission technology cannot effectively utilize the 5G throughput according to an environment. In this paper, we propose a method to predict the video transmission throughput by using the location information and the communication logging information. The logging information including past throughput with the location is created by a UE. A UE refer to the logging information for prediction when the BS which is attached to the UE is changed or the radio quality is declined. We developed a video transceiver implemented the proposed method and evaluated its performance. We show that the proposed method can stably transmit high resolution video compared to the conventional method.